An Energy-Efficient And Reliable Asynchronous Messaging Framework For Multi-Core Socs

In the modern era, multi-core technology is one of the effective communication between the various IP-enabled cores and it is a critical challenge. As more and more IP cores are integrated into multi-core system-on-chip (M-SoC), traditional bus architectures are inadequate to meet the performance demands of modern systems.  Also, interconnection architecture for on-chip multi-core communication has an increasing presence in research due to its scalability and efficient communication. M-SoC is the integration of multiple processing cores to implement the communication performance effectively within the SoC. Consequently, conventional messaging protocols fail to optimize the clock distribution challenges and imperfect fault tolerance. To achieve better performance, adaptive routing algorithms need to monitor network status and make routing decisions informed by congestion data, which unavoidably increases design complexity and overhead. Therefore, this paper developed an Adaptive Fault Tolerant with Hybrid Congestion Aware routing protocol (AFT-HCARP) is designed to improve the performance while ensuring the asynchronous communication protocol. Here, a deterministic-adaptive hybrid routing strategy was utilized to reroute the transmitted messages around the overloaded modes. In addition, the proposed AFT-HCARP protocol is integrated with dynamic routing adaptation as well as a destination-based adaptive routing algorithm to validate the finest message transferring under several network conditions, which are varied under the fault scenarios. Moreover, the experimental outcomes were verified in terms of throughput performance under various fault conditions, latency to load, and finally comparative analysis is performed across the baseline protocols. Furthermore, the results show that developed AFT-HCARP reduces average packet delay as 3.8cycles and enhances the throughput as 0/0098 flit/node/cycle. From the comparison, fault recovery time is 4 cycles, and packet delivery ratio is 97.60% than the conventional routing schemes.